teras.preprocessing.GAINDataTransformer#
- class teras.preprocessing.GAINDataTransformer(categorical_features=None, continuous_features=None)[source]#
GAINDataTransformer class that performs the required transformations on the raw dataset required by the GAIN architecture.
- Parameters:
categorical_features (
Union[List[str],Tuple[str]]) – list, List of categorical features names in the dataset. Categorical features are encoded by ordinal encoder method. And then MinMax normalization is applied.continuous_features (
Union[List[str],Tuple[str]]) – list, List of numerical features names in the dataset. Numerical features are encoded using MinMax normalization.
Methods
__init__([categorical_features, ...])fit(x)fit_transform(x)get_metadata()inverse_transform(x)Inverse Transforms the transformed data.
load(filename)Loads the saved state of DataTransformer from the json file.
save(filename)Saves the fitted state of DataTransformer instance for portability, in the json format.
transform(x)Transforms the data (applying normalization etc) and returns a tensorflow dataset.
Attributes
metadata